Conventional systems which provide item fit recommendations (i.e., a prediction regarding which size of a particular item may best fit a particular consumer) rely on detailed technical data on items, such as item dimensions and/or other attributes (e.g., waist circumference, inseam, etc.) Although this technical data can be helpful in making fit recommendations if available, often it is not available for use. For example, often the suppliers of particular items consider technical data on those items proprietary, and may not wish to reveal the technical data to another party for fear that the items may be replicated.